node-connect skill
Diagnose OpenClaw Control UI browser and native Android, iOS, or macOS node connection failures across route, auth, pairing, QR/setup-code, and reconnect states.
Is the node-connect skill safe?
Clean: nothing in its files matched our rules. We read 1 file in the folder on 2026-10-25.
No findings.
Install the node-connect skill
A skill is a folder. Copy it into your agent's skills folder and the agent loads it when the task matches its description.
git clone --depth 1 https://github.com/openclaw/openclaw.git /tmp/openclaw mkdir -p ~/.claude/skills cp -r /tmp/openclaw/skills/node-connect ~/.claude/skills/node-connect
In the Claude apps, zip the folder and upload it from the Skills settings. The folder on GitHub
The instructions your agent would load
SKILL.md as published, without the frontmatter. Read it on GitHub
Node Connect
Goal: fix one exact client against one exact Gateway, then prove that client's fresh connection.
1. Lock the target
Record the target environment/profile, OpenClaw binary, config/state root, Gateway URL/port, and service before changing anything.
- Use the named deployment wrapper/profile for every config, log, service, device, and node command.
- Never drift to a bare openclaw, proof environment, or similarly named deployment.
- If the global executable is stale, invoke the target through its owner.
- Verify status, config, logs, and process identity describe the same Gateway.
Identify the target Gateway before changing pairing or auth.
2. Classify the client
Classify from the request and Gateway log before choosing commands:
- Control UI browser: the user says browser, dashboard, Control UI, or webchat; logs show client=openclaw-control-ui or mode=webchat.
- Native mobile/node: the official app shows Connect, Scan QR, or setup code; logs/request metadata show a native client or role=node.
A phone can be either client. Do not use openclaw qr or openclaw nodes status for a phone browser; those belong to native mobile/node pairing.
3. Observe the failing attempt
Run these through the locked target:
openclaw gateway status --deep
openclaw logs --follow --json
openclaw devices list
openclaw config get gateway.mode
openclaw config get gateway.bind
openclaw config get gateway.remote.url
openclaw config get gateway.auth.mode
openclaw config get gateway.auth.allowTailscale
openclaw config get gateway.tailscale.modeHave the client retry once while logs are live. Correlate its client ID, mode, platform, address, auth result, device ID, user, and close code. Ignore other paired devices.
Interpret the first failed transition:
- No matching Gateway attempt: route, DNS, TLS, origin, or proxy problem.
- tokenmissing, passwordmissing, mismatch, or Tailscale identity failure: auth failed before pairing, so an empty pending list is expected.
- pairing required: route and auth succeeded; approve the exact pending request.
- client connected: match conn, client, mode, and role to the exact client. Admission precedes the hello response; if followed by 1006, preserve auth/pairing and inspect hello delivery, lifecycle, transport, proxy, or reconnect. Older builds use overlapping authenticated user connected / webchat connected records that omit ephemeral clients.
- 1006 without a matching application-level admission record does not prove auth/pairing; inspect the earlier handshake transition.
4. Prove the route
Choose one topology: same machine, LAN, tailnet, or public reverse proxy. Do not mix them.
- Browser Control UI needs HTTPS or localhost for browser device identity. A remote plain-HTTP Tailnet/LAN URL is not a valid substitute.
- gateway.tailscale.mode=off means OpenClaw is not managing Serve/Funnel. It does not prove that Tailscale or an externally managed Serve route is absent.
- When Tailscale is involved, inspect live state rather than inferring it from OpenClaw config:
tailscale status --json
tailscale serve status --jsonMatch the client's URL to the listener/proxy route reaching the locked Gateway.
5A. Control UI browser lane
Restore browser auth before looking for a pairing request:
- For token/password auth, enter the credential in Control UI settings. Never put permanent secrets in chat, logs, or URLs.
- Prefer openclaw dashboard on the Gateway host for a one-time signed handoff. Use --no-open only when the operator can retrieve that host's clipboard, and keep the host browser/clipboard outside agent tooling. Never capture dashboard --json: it can expose the handoff and shared credentials. Never relay, rewrite, or send a loopback handoff URL to a remote phone.
- For Tailscale Serve, verify the live route and forwarded identity. Enable gateway.auth.allowTailscale only for that intended trust boundary. Verified Tailscale Control UI auth with browser device identity can skip pairing.
After auth succeeds:
- Retry; never infer pairing from the pre-auth attempt. Token/password auth can reveal pairing required; verified Tailscale identity can skip it.
- If logs say pairing required, re-list devices and approve the exact request ID.
- A successful verified-Tailscale connection may correctly create no pending or paired-device row.
- After a repeated 1006, preserve auth/pairing and inspect reconnect evidence.
5B. Native mobile/node lane
Inspect the native route through the locked target without exposing the setup credential:
openclaw qr --json | jq '{gatewayUrl, gatewayUrls, auth, access, accessDowngraded, urlSource}'For a CLI controlling a remote Gateway, add --remote before --json; it selects gateway.remote.url and remote credentials. If the redaction filter is unavailable, do not run raw QR JSON in agent-visible output.
Verify gatewayUrl and urlSource. The setup code is password-equivalent: have the operator copy it from Control UI → Devices → Pair device, or run openclaw qr --setup-code-only in a terminal outside agent tooling and paste it directly into the official app. Never relay it through agent/chat/tool output. Generate a fresh code after a URL/auth fix or expiry.
If the app reports pairing required:
openclaw devices list
openclaw devices approve --latest # preview only; exits without approval
openclaw devices approve <requestId>
openclaw nodes status--latest only previews the current request; never treat it as approval. Re-list immediately before the exact-ID command because retries can supersede the request. Never approve by position, age, or similarity.
6. Correlate and finish
Before approval, match available request, device/public-key, client, mode/role, platform, address, user, and retry-time facts.
Declare success only after a new attempt made after the final change proves all applicable checks:
- the exact client reaches the locked Gateway;
- intended auth succeeds;
- the browser completes initial requests, or the native node appears in openclaw nodes status;
- approval used the exact request ID; and
- no immediate auth, pairing, or reconnect failure follows.
A QR/setup code, launched browser, empty pending list, approval, paired-device count, or Tailscale ping proves only one transition.
Report the diagnosis, chosen route/auth lane, exact-client evidence, and any remaining failed transition.
More skills from openclaw/openclaw
- A1passwordSet up and use 1Password CLI for sign-in, desktop integration, and reading or injecting secrets.
- Aacp-routerRoute plain-language requests for Claude Code, Cursor, Copilot, OpenClaw ACP, OpenCode, Gemini CLI, Grok Build, Qwen, Kiro, Kimi, iFlow, Factory Droid, Kilocode, or explicit ACP harness work into either OpenClaw ACP runtime sessions or direct acpx-driven sessions ("telephone game" flow). For coding-agent thread requests, read this skill first, then use only `sessions_spawn` for thread creation. Codex chat binding defaults to the native Codex app-server plugin unless ACP is explicit or background spawn needs ACP.
- Aadd-model-providerAdd and live-prove a model provider with non-interactive config one-liners, without exposing credentials.
- Aagent-transcriptRequested GitHub PR/issue agent transcripts: redact, trim, preview, and insert safely.
- Aapple-notesCreate, view, edit, delete, search, move, or export Apple Notes via the memo CLI on macOS.
- Aapple-remindersList, add, edit, complete, or delete Apple Reminders and reminder lists via remindctl.
- Aauto-qaRun OpenClaw-wide autonomous QA and live/stress campaigns across independent subsystem lanes, with verified fixes and a resumable evidence report.
- AautoreviewStructured code review through the shared OpenClaw agent-skills installation.
- Abear-notesCreate, search, and manage Bear notes via grizzly CLI.
- Ablock-kitUse proactively for structured or interactive Slack replies, and when asked to author or validate native Slack Block Kit JSON.
- AblogwatcherMonitor blogs and RSS/Atom feeds for updates using the blogwatcher CLI.
- AblucliBluOS CLI (blu) for discovery, playback, grouping, and volume.